Corrosion Science, Vol. 38, No. 6, pp. 999-1002, 1996 Copyright 0 1996 Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 001IS938X/96$[5.00+0.00

Pergamon

PII: SOOlO-938X(%)08036-4 SHORT COMMUNICATION THE PITTING SUSCEPTIBILHY OF STAINLESS STEELS AFTER A STAY AT OPEN CIRCUIT IN ACIDIFIED CHLORIDE CONTAINING SOLUTION C. BOULLERET, *D. GORSE, ‘B. BAROUX CECM-CNRS, 15 rue Georges Urbain, 94407, Vitry/Seine, France ‘Centre de Recherche d’Ugine, 73400, Ugine, France

m - In acidic chloride-containing solution, the passive fti formed at open circuit becomes unstable over a cetin period, as revealed by subsequent potentiostatic electrochemical noise measurements, showing specifx Cutrentpeaks. A hypothesis is made about the possible influence of a salt layer forming on pit walls during both the propagation and repassivation of unstable pits.

The pitting corrosion resistance of stainless steels is known to generally improve after open circuit potential (Voc) aging in the corrosive solution itself, provided that no stable pitting occurs during this stay at V,,‘. passive film formed at V,

This paper shows that, in an acidified chloride-containing

solution, the

over a certain aging period appears to become unstable when imposing

further anodic polarization, whereas a somewhat longer or shorter aging time at open circuit potential would be sufficient to hinder the development of such electrochemical instabilities. It is noticeable that this unstability period occurs after about 15 min. at VW, which is a standard time delay imposed before starting potentiostatic studies of pitting corrosion. Specimens were cut from AISI 430 type industrial sheets (thickness lmm, annealed condition). Disks of 0.785cm’ area were carefully2 mechanically diamond-polished, then left 24 h in air before testing, last allowed to stay at Voc in an acidic (pH 3) de-aerated chloride containing-solution

(NaCl

OSM) up to 2 h. Open circuit potential versus time plots are remarkably reproducible (Fig. I), at least as concerns the average Voc value. Within the 2 min. following immersion, the average open circuit potential falls off abruptly, its value attaining rapidly (in less than 5 s) a minimum located in the range of hydrogen evolution; then after a turning point, Voc begins to re-increase linearly at a very slow rate. Before the average VW drop, fluctuations revealing metastable pitting are visible. The nearly symmetrical shape and size (- 8 s) of these potential transients resemble those already observed by Pistorius3. In an individual transient, the potential decrease corresponds to metastable pit growth, at least a part of the growth charge being stored by the surrounding surface4, the other part being supplied by the farad& cathodic current. After repassivation (occurring~for the potential minimum value), potential growth results from the release of the previously stored charge. After the average Voc drop, no potential fluctuations are observed. It does not mean that the steel surface no longer undergoes metastable pitting, but simply that the potential is controlled by hydrogen evolution, and that no capacity charge or discharge may occur. Manuscript

received 21 December

1995; in amended 999

form 5 March

1996.

1000

Short commumcauon

O-

Instability Period

g Y? > 5 +

0.1

_

o*,

”

”

“.

‘-

8

0

1000

3000 Time

7000

5000 (s)

Fig. 1. Typical plot of the open circuit potential as a function of the time of exposure in 0.5 M NaCl at pH 3, for an AISI 430 type stainless Steel. The various ageing times preceding the potentiostatic tests conducted at +20 mV(SCE) are indicated, with the main result of the tests : development of a stable pit (A), or not (*) (Inset : potential fluctuations occurring before the average VW drop).

After various lengths of time @c (from 20 s to 2h), the potential is stepped to a test value Vp less than the pitting potential, Vpit, approaching +50mV(SCE) after 2h at open circuit. The unfiltered current fluctuations are measured using a PAR 273 potentiostat. The results obtained at +20 mV(SCE) are presented below, but these findings were confirmed by all tests conducted at higher VP ( 1. The anodic transient appears often somewhat truncated. The current decay occurs also in at least two steps : a first one during which the current decreases to a level slightly

I . . . I . 158 156 Time (s)

.

. 160

Fig. 4. A typical me&stable event observed inside the Instability Period (1500 s at VW) alter 48 mm. polarization time at +2OmV(SCE) (‘tc = 0.17 s, Qc = 3.7 nC).

1002

Short communication

I

I

,; 14 : 12

0.4

jl0

+

(0

0

-8 0.3

1 -6

8

*4

Time at Voc (s) Fig. 5. Diagram showing the maximum decay time (+) and the corresponding charge (0) passed during the cathodic transients obtained in 0.5 M NaCl, pH 3, during the Instability Period (hatched area), by comparison with the same quantities found outside that period.

above the base line continued by a sudden fall off of a few tens of nA below the base line terminated by a re-increase

to the base line taking between 0.2 s and about 0.6 s. The duration and charge carried

in the cathodic transients are higher than those measured outside the instability period (Fig. 5). To obtain the present results, the following conditions must be filled : first a certain aging time at Voc,

then a certain polarization

existence

of an Instability

(concerning

time, and last a sufficiently

corrosive

solution

Period detected together with a staircase shape of the current transients

both the rise and decline stage) suggests

a possible

role played by a thin salt film’

forming on the walls of the pit cavity. This salt film would not completely

dissolve during the life of

the unstable pits leading to an additional resistance hindering both pit propagation The collapse

(Fig. 5). The

of this salt film would be followed

terminated by a decline in one step corresponding

and repassivation.

by a clearly visible surge of the anodic current, to a sudden repassivation.

The hypothesis

could be

made that the cathodic transients correspond to the removal of the charge stored in the so modified pit walls. However, one notes that the charge released is small, by comparison

with the amount carried

in the anodic transient (by a factor of at least 50). REFERENCES 1. 2. 3. 4. 5. 6. I.

B. Baroux, in “Corrosion Mechanisms in Theory and Practice”, P. Marcus and J. Oudar Eds, M. Dekker Inc., New-York, pp. 265-309 (1995). B. Boulleret, J.L. Pastol, J. Bigot, B. Baroux, D. Gorse, in proceedings of the 2nd International Sympcsium on Ultra High Purity Metals, St Etienne, June 13-16 (1995). P.C. Pistorius, ASTM International Symposium on Electrochemical Noise measurements for Corrosion Applications, Montreal, May 16-17 (1994). H.S. Isaacs and Y. Ishikawa, J. Electrochem Sot. 132, 1288 (1985). D. Gorse, C. Boulleret, B. Baroux, ASTM International Symposium on Electrochemical Noise measurements for Corrosion Applications, Montreal, May 16-17 (1994). D.E. Williams, J. Stewart, P.H. Balkwill, Corros. Sci. 36 1213 (1994). G.S.Frankel, L. Stockert, F. Hunkeler, H. Boehni, Corrosion 43,429 (1987).